How to solve theERROR invalid device function? Please

[zhuheng96]

Basic environment info:Jetson Orin NX, Ubuntu 20.04, Jetpack5.1.1, python 3.8, pytorch 1.11.0, CUDA 11.4.315.
I run a python code which uses pytorch module and yolov5.
The error as following shows:
Traceback (most recent call last):
File “main.py”, line 80, in
bboxes = detector.detect(im)
File “/home/algo_test/unbox_yolov5_deepsort_counting-main/detector.py”, line 47, in detect
pred = self.m(img, augment=False)[0]
File “/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py”, line 1110, in _call_impl
return forward_call(*input, **kwargs)
File “/home/algo_test/unbox_yolov5_deepsort_counting-main/models/yolo.py”, line 123, in forward
return self.forward_once(x, profile) # single-scale inference, train
File “/home/algo_test/unbox_yolov5_deepsort_counting-main/models/yolo.py”, line 139, in forward_once
x = m(x) # run
File “/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py”, line 1110, in _call_impl
return forward_call(*input, **kwargs)
File “/home/algo_test/unbox_yolov5_deepsort_counting-main/models/common.py”, line 45, in fuseforward
return self.act(self.conv(x))
File “/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py”, line 1110, in _call_impl
return forward_call(*input, **kwargs)
File “/usr/local/lib/python3.8/dist-packages/torch/nn/modules/conv.py”, line 447, in forward
return self._conv_forward(input, self.weight, self.bias)
File “/usr/local/lib/python3.8/dist-packages/torch/nn/modules/conv.py”, line 443, in _conv_forward
return F.conv2d(input, weight, bias, self.stride,
RuntimeError: :0: cudaFuncSetAttribute(kernel_entry, cudaFuncAttributeMaxDynamicSharedMemorySize, integer_cast<int32_t>(launch_configs[0].smemSizeInBytes)): invalid device function

[AastaLLL]

Hi,

Have you tried to run the model on Jetson before?
Could you check with the author to see if the sample has any dependencies on the hardware (ex. GPU architecture or memory usage?)

Thanks.